Foundry, a new player in the tech industry, has launched a groundbreaking GPU-based cloud platform designed specifically for AI workloads. After securing $80 million in funding earlier this year, Foundry aims to significantly reduce both costs and complexity for AI teams.
The Foundry Cloud Platform offers a unique approach to GPU usage. It allows users to reserve GPU capacity for short periods, from a few hours to a full day. This system not only cuts costs but also increases efficiency. Additionally, the platform supports spot instances for workloads that can tolerate interruptions.
Foundry’s platform includes automated Kubernetes workload orchestration. This feature removes the need for manual scheduling, enabling AI teams to focus more on innovation and less on infrastructure management.
Early adopters, such as the Arc Institute and AI startup Infinite Monkey, have already experienced the benefits of Foundry’s platform. “Foundry’s platform gives us the computing power we need, exactly when we need it,” said Patrick Hsu, CEO of Arc Institute.
Jared Quincy Davis, Foundry’s CEO and Founder, was inspired by his time at Google’s DeepMind, where he observed the development of AlphaGo. In 2022, he began building Foundry with the goal of creating a cloud service tailored to AI’s unique needs. With the launch of the Foundry Cloud Platform, his vision has become a reality.
By addressing inefficiencies in the GPU market, Foundry is set to play a critical role in advancing AI development and making new technologies accessible to innovators.